This disclosure generally relates to online systems, and more specifically to predictions of household features of a user, e.g., size of the household and demographic composition of the household, based on image data associated with the user.
Content providers produce content that is targeted to certain audiences within online systems. Users interact with content received from the online systems. With the advent of online systems, such as social networking systems, content providers have increasingly relied on the online systems to effectively deliver content to their target users in target households. However, existing solutions of content delivery to a target household are not effective. For example, the content providers produce and send the content items to the target household without knowing the size of the target household and the demographic composition of the household (e.g., number of members in the household, characteristics of the household members, shared interests of the household members, etc.) Furthermore, the content providers often have no knowledge of what electronic devices are used by the household users to access content provided by the content providers, how the electronic devices in the household are associated with the household users, and whether the household users associated with the electronic devices in the household are users of the online system that receive content from the content providers. Without such knowledge of a user's household features, most of content items that are sent to the user are poorly tailored to the user and are likely ignored by the user. As a result, the content providers waste resources while sending the content items to the user.